Microwave tomographic imaging of the breast has successfully advanced to initial clinical testing for monitoring treatment response to breast cancer neoadjuvant chemotherapy. The Dartmouth microwave imaging team has performed over 500 breast exams in both diagnostic and therapy monitoring settings which have demonstrated that recovered images of the endogenous dielectric properties are highly specific to distinguishing tumor from normal tissue, and to following response to therapy. In initial therapy monitoring patient studies, results indicate that images acquired as early as 30 days after the start of treatment correlate well with pathological outcomes. In this project, the Dartmouth microwave imaging group will team with General Electric's (GE's) Global Research Center to further refine and optimize the current technology as an important step towards translating it into the clinic. Design modifications are conceived to advance the technology using best engineering practices and to prepare it for a two center imaging study. Because the current system utilizes a modest amount of measurement data and has the lowest computational requirements of any microwave tomographic system world-wide, the overall costs will remain low and system configuration simplicity will be preserved. Designing for a second site will require a significant level of additional rigor be applied to the current system to ensure that it an be operated reliably and efficiently at a remote site. Refinements will incorporate previously proven spatial prior techniques designed to isolate the dielectric property recovery in pre-determined zones - specifically in tumor, and fibroglandular and adipose tissues without unduly biasing the final images. In addition, multi-frequency imaging methods will be refined and added to increase the spectral content of the recovered data and subsequently to improve the specificity of the reconstruction process. Each of these innovations will be incorporated into the new compact and portable design where special attention has been paid to clinical workflow requirements.